Debunking Dividend Myths: Part 1
This post is the first in a series exploring common myths and misunderstandings about dividend investing. The goal of the series is to argue that many investors following a dividend-focused strategy may be better off with broad-based index funds.
Dividend Myth #1: Companies that pay dividends are inherently better investments than those that don’t.
Many investors take it for granted that dividend-paying companies are superior to those that do not pay a yield. But this idea has been the subject of debate for decades, and many academics believe that it is irrational.
Let’s start with something everyone can agree on. Equity returns have two components: capital gains (price increases) and dividends. Add them together and you have the total return for a stock. Ignoring taxes and transactions costs, a stock that pays no dividend but increases in price by 6% provides precisely the same return as one whose share price rises 4% and pays a 2% dividend.
Dividends lead to a drop in share price
What many investors don’t grasp is the direct relationship between share prices and cash dividends. If a company’s stock is trading at $20 and it pays a $1 dividend, its share price will fall to $19 on the ex-dividend date. This price drop may not be penny for penny, because it will be combined with normal fluctuations in the daily markets. But there is always a trade-off. The failure to understand this point is the reason that so many investors think of dividends as “free money.”
In 1961, Merton Miller and Frank Modigliani published a landmark paper that became the basis for what is now known as the dividend irrelevance theory. They argued that whether or not a company pays dividends should not matter to shareholders, because it does not affect their overall returns. Dividend policy simply determines whether investors end up with a share valued at $20, or a share worth $19 plus $1 in cash.
Not everyone accepts the Miller-Modigliani model. Two other economists, Myron Gordon and John Lintner published a counter argument that has come to be called the bird-in-the-hand theory. Their idea was that shareholders cannot be sure that a company will spend its capital wisely, so a dollar paid in dividends is preferable to one kept as retained earnings. This may be true, but investors who advocate DRIP programs cannot logically subscribe to this theory: if they truly believed it, they would not reinvest their cash dividends in new shares.
The illusion of income v. capital
Why do shareholders believe so strongly that a $1 dividend is preferable to a $1 capital gain? Meir Statman looked at this question in a 1984 article called “Explaining Investor Preference for Cash Dividends,” coauthored by Hersh Sheffrin. He also reviews the idea in his new book, What Investors Really Want, pointing out that receiving $1,000 in dividends is no different from selling $1,000 worth of stock to create a “homemade dividend.”
Even when this idea is explained to people, most refuse to accept it. Statman suggests that it comes down to a cognitive bias called mental accounting. Investors categorize $1,000 in dividends as income that they will happily spend, but the idea of selling $1,000 worth of stock is “dipping into capital,” which causes them great anxiety. This idea is deeply ingrained in many investors, but it is an illusion, because a company that pays a dividend to shareholders is depleting its own capital.
Other sources of shareholder return
There are other compelling arguments against the inherent superiority of dividend-paying stocks:
- Many companies that pay no dividends use their free cash to repurchase stock. These buybacks increase the value of the remaining shares, which benefits all shareholders. There is no logical reason to prefer dividends over stock repurchases when the net benefit to investors is the same. (Indeed, share repurchases have no tax consequences, while dividend payouts do.)
- Profitable companies that have not reached maturity can often earn much higher return on equity by reinvesting their earnings in growing businesses. Many highly profitable companies (especially in technology and mining) do not pay dividends because they can put that cash to much more productive use. These companies will typically begin paying dividends later in their life cycle, when those growth opportunities disappear.
- It has long been known that the over the long term, small-cap stocks have dramatically outperformed large-caps (with correspondingly more risk). The small-cap premium wasmore than 3% during the past 80 years. Most small companies do not pay any yield, so investors who select only dividend-paying stocks are ignoring the segment of the equity market that has enjoyed the highest long-term returns.
Investors who use broad-based index funds accept that equity returns come from profitable businesses, some of which pay cash dividends, and some of which put their earnings to work in other equally rewarding ways. In the end, as Meir Statman writes, “a dollar labeled dividends is as green as a dollar labeled capital, so rational investors are indifferent between the two.”
Other posts in this series:
Dividend Myth #2: Dividend investors are successful because they select excellent companies and buy them when they are attractively priced.
Dividend Myth #3: Dividend-paying stocks are a substitute for bonds in an income-oriented portfolio.
Dividend Myth #4: You can beat the market with common sense: just focus on blue-chip companies with a competitive advantage and a history of paying dividends.
Dividend Myth #5: It’s easy to build a well diversified portfolio of Canadian dividend stocks.
{ 26 comments… read them below or add one }
http://img338.imageshack.us/img338/614/dividend.png
The graph was generated by Kenneth French who is a professor of finance at the Dartmouth college. The graph shows the performance of 4 different portfolios. The blue line represent the S&P TSX index. The orange line tracked stocks who did not pay dividends at all. The purple line tracked stocks that paid the lowest 30% of dividend yields, and the green line tracked stocks in the highest 30% yield group. Portfolios were rebalanced every year.
For the last 30 years investing in high yielding dividend paying stocks clearly outperformed. Growth stock “paying no dividends” performed poorly.
@Germack: This graph appears in many discussions of dividend investing. Two observations: these returns are hypothetical and assume that the entire portfolio is changed once a year based on the current yield of the stocks. This is an extraordinarily active strategy. Dividend-growth investors, who are mostly buy-and-holders, don’t use a strategy even remotely like the one suggested by this graph, so it cannot be used to support their argument.
Second, Kenneth French is co-creator of the Fama-French Three Factor Model, which explained that returns can be almost entirely explained by market exposure, value characteristics and company size. French is one of the leading minds behind Dimensional Fund Advisors, a firm whose funds are designed to passively capture the returns of entire asset classes. He has been an advocate of globally diversified passive portfolios for decades and has never recommended a dividend-focused strategy based on these data.
The graph shows the return from the value premium and not the dividend premium. The three ingredients in the the Fama-French model are market, small and value, not dividends. In fact, Jim Davis of Dimensional Fund Advisors did a paper which shows sorting by dividends is the worst way to access this premium.
The graph is not showing the relationship between dividends and returns. It is showing the relationship between low price (a stock’s credit score) and returns. It just so happens that dividend stocks have low prices, but when a portfolio is built of dividends it is the worst way to access the value premium. Nice graph, wrong interpretation.
RE: growth companies v. dividend payers. Let’s say Disney has an ROI of 12% on their stock from making movies for my kids. (Sometimes they will make $25 bucks a year sometimes only $3 bucks but the average ROI is 12%.) If they earn 10 bucks from making movies and I ask them to pay me 6 bucks as a dividend they are left with $4 to make movies. They need the $10 to get the 12% ROI. The more I make them pay me out of their earnings, the less they will have to make movies. It is simple math.
The beauty of an index is I have hundreds of Disneys making earnings, which takes away my unfounded fear that they will buy hookers, coke and fly around in private jets. The average ROI on indexes is around 10 to 12%. Do I want to take that money out to invest in 2% bonds or leave it in to grow?
I strongly disagree with this paragraph:
“Many companies that pay no dividends use their free cash to repurchase stock. These buybacks increase the value of the remaining shares, which benefits all shareholders. There is no logical reason to prefer dividends over stock repurchases when the net benefit to investors is the same.”
Although in theory what you’re saying is true (in fact, given Canadian tax laws, stock repurchases are preferable), in practice it is far from true. The majority of stock buybacks are necessary to combat dilution from insider compensation, and occur at times appropriate to that, rather than at times when the company stock is priced attractively. It’s quite sobering to look at the stock repurchase programs of many of the members of the S&P 500 and realize that over the course of that same year, the number of outstanding shares actually increases, despite the buybacks.
While I’m all for compensating insiders adequately, even lavishly, in many cases the sheer scale of what’s happening becomes unfair to shareholders. To discourage this, it is rational to prefer dividends over repurchases. (A history of dividend increases creates an expectation of future dividend increases, which acts as a restraining function on management.)
I agree with your points here and don’t believe that dividend paying stocks are “better” than stocks that don’t pay a dividend. I also don’t believe that a dividend is preferred to a capital gain.
Steve brought up some very interesting points yesterday about hindsight bias and how to pick winning stocks over the next 20 years rather than looking at the last 20 years. That will be a challenge for sure. However if you don’t use a DRIP (I don’t) you will continue to accumulate cash to either add to existing positions when they are value priced, or to find a new dividend paying stock that meets your criteria (yield, p/e, Graham, etc).
Where I struggle with indexing, or even with investing in non-dividend paying stocks, is what do you do in retirement? What happens when the market crashes like it did in 2008-2009 and you lose 40% of your portfolio when you are set to retire? At some point you need to take your money out, or switch to an annuity or fixed income product that erodes over time due to inflation.
I’m looking to build income for retirement, preferrably growing income. And while it’s true that dividend paying stocks aren’t immune to market crashes, reductions in dividends are rare occurances. Wouldn’t you be happy if your income rose by 8% in retirement during a bear market?
There’s no reason to believe that you will do better with dividend paying stocks than non-dividend paying stocks. Since we cannot predict in advance which stock and bond asset classes will perform the best, we DIVERSIFY.
BUT: I view dividend-paying stocks as a separate asset class. I prefer investing in high-dividend stocks (or with an ETF), which is a mild form of value investing, than in high-yield bonds.
Dividend investing is a strategy for the long haul. The major attraction is not to make money from price increases, it is the dividend itself. Coca-Cola (KO): 2001-12: $47 and $0.72 div. / 2009-12: $57 and $ 1.64 div. The yield on cost is 3.5 % from 1.5 % .
According to Ned Davis Research, based on the S&P 500 from 1972 to 2006, non-dividend paying stocks produced an annual average return of 4.1% vs dividend-paying stocks ’10.1%.
One of the most cited reasons for inferiority of dividends is that shareholders are better off if the company uses that money to buy back shares. Often companies pay top dollar for their shares. They don’t buy low. A study from S&P shows that very few companies repurchased their shares in 2002, the bottom of the bear market. Buybacks have peaked in 2007 simultaneously with the market’s peak.
Lehman Brothers had authorized the buyback of up to 100 million shares on 2007-0
There is a strong case for considering dividend-paying stocks as an asset class in any portfolio.
This is great stuff Dan. I agree with Chris above that in practice dividends are preferable to share buybacks. In terms of asset allocation of dividend paying stocks vs. broad market index funds, the most important factor is one’s proximity to retirement. In retirement, I think dividend stocks (which can be acquired via CDZ or XDV) are preferable to broad based funds (like XIC) because a greater proportion of the return is in cash, which means downside protection. During retirement (65 – ?) my IPS calls for 20% exposure to equities, which will be split evenly between a Canadian dividend ETF and US one (such as DVY).
@ART: The graph shows dividend yield not value premium.
First of all, I like the topic and discussion. Thanks for putting this series together knowing full well it was going to be a hot topic!
I agree that your myth is just a myth. There could always be found an argument to counter any investment strategies. It really depends on what you are comfortable with. For me, dividend investing is more predictable with a return I am happy with. Compound growth is what I am banking on. It’s an exponential growth rather than a linear growth of a market. It doesn’t mean that I buy and hold forever … I sell and take profits and buy better opportunities based on the market conditions of different companies. I don’t chase the yield but I look for companies that have a good yield with a predictable (as much as it’s possible to predict) growth in dividends.
It’s funny how in many cases the ex-dividend dates sees an increase rather than a drop of value for the dividend. Theoretically, you are right but the herd isn’t following that mantra…
Most of what has been said is true, BUT, to me it’s all about getting dividend checks in the mail, on a monthly basis. And, after 18 years of adopting the dividend strategy, my chosen stocks have grown considerably, plus my yield on initial investment has grown exponentialy.
@Passive Income Earner
“Compound growth is what I am banking on. It’s an exponential growth rather than a linear growth of a market.”
The market growth is also exponential, the capital gains from previous years is generating capital gains from this year onwards. Wouldn’t that make the indexing strategy an exponential growth one as well?
This is an interesting series and discussions that I will be following along for sure.
I believe that an investment should pay a regular income. Not many of us can afford to put our money in a growth vehicle that pays no benefit until it is sold. It would be ludicrous for me to sell my stock 5% at a time in lieu of a 5% dividend,
because the brokerage fees would put as much money in the stockbrokers pocket
as in mine .
Thanks to everyone for such an open-minded discussion.
@Mark: Exactly right. I’ll deal with this in a future post, but there is a persistent idea that reinvesting dividends is different from all other forms of compounding. A stock that grows at 6% compounds precisely the same way as one that grows at 4% and pays a 2% dividend that is reinvested. (Except that the dividend-paying stock incurs more tax.)
@Wendell: I acknowledge that investors who are regularly drawing on their portfolio for income can make an excellent case for dividends. But for most of our investment life we are in the accumulation stage and do not typically need the income. Even in retirement, the practical problems you describe are easily overcome and should not dictate an entire strategy. In a discount brokerage account, trades typically cost $10, plus a small bid-ask spread — these costs are almost negligible. There are also ways to set up systematic withdrawal plans to reduce the costs even further.
@Mark
Going back 80 years, the market growth does look exponential. I remember the graphs you see in a banks when I sign the mortgage papers but I felt the exponential factor wasn’t as steep as what the dividend growth and compound growth can give you. More so when you look at sub-set of the markets at which point it’s more linear over a 10, 15 year period. Again, it really depends on which investment you hold and the dividend increase you can get. It’s very hard to compare.
I believe in index investing and recommend it to many. It’s what I will recommend to anyone starting. I send them right over to your site!
Personally, I just believe I will do better with my strategy which includes dividend and growth. I will also take my profits when appropriate. The jury is still out though … I have to let the numbers speak for themselves and not hypothetical portfolios from research and book writers. I need to find a way to compare my portfolio to the S&P index (my broker doesn’t provide that), or any investments for that matter for a comparative purpose, I have not done it yet.
Good on you Dan to take this subject on, somewhat controversial as it may be
IMO, dividend-paying companies are not necessarily superior, rather they provide me with assurance about their long-term prospects and consequently pay me for being an owner of their business. Do I avoid RIM because it doesn’t pay a dividend? For the most part, yes. Why? I feel that dividends never lie. In order for a company to pay a dividend, a consistent one at that, a company must have the money to pay it. Sure, said company could also buy back shares, acquire new companies or other lines of business with their capital but as a shareholder I want to get paid. Your arguement about income vs. capital is understandable: receiving $1,000 in dividends is no different from selling $1,000 worth of stock to create a “homemade dividend”. While true, I am losing-out on my ability for existing shares to buy more shares via dividend reinvestment if I performed the latter transaction. To this end, I too am banking on the magic of compounding.
I would argue an investor would do quite well to own half of the stocks in XDV or CDZ outright, never paying a management fee ever again.
For many reasons, indexing works very well because in part what you touched upon above: “over the long term, small-cap stocks have dramatically outperformed large-caps”. More risk = more reward. Index investors are rewarded with all types of equities, large-cap, medium-cap and small-cap stocks. Most small companies do not pay any dividends so you’re right, investors who select only dividend-paying stocks are ignoring this equity segment. That said, I’ll take my chances with a domestic powerhouse like a Canadian bank or an international congomerate like Coca-Cola than a junior mining stock on the TSX. With the former, I don’t have to worry about when or if I will get paid. This is why dividend investing works for me, with a few good stocks, I have some level of assurance where my income is coming from regardless of Mr. Market’s behaviour.
A dividend-investor owning stocks that pay consistent dividends never really has to worry about Mr. Market’s attitude, an equity index investor should, especially in retirement.
I look forward to more comments and posts on your dividend investing theme Dan.
Cheers,
Mark
I am somewhat confused with the renewed focus on dividend indexes as a deviation from the pure passive index investing. While this is second article in a row (and apparently not the last) describing why dividend-based investing is problematic, a new model porfolio has been added for 2011 (Yield Hugry) and it has both CDZ and XDV as its equity components. Which is it – should ETFs such as CDZ or XDV be used as part of a self-respecting, card-carrying couch potato portfolio or not? I am asking in the context of a portfolio where the horizon is short to medium (3-5 years), not for long term investing (say, RRSP with 20 years horizon).
@Dejan: Thanks for the comment, and sorry if my position wasn’t clear. The Yield Hungry Portfolio is a suggestion for income-oriented investors who are withdrawing and spending cash from their portfolio, and it’s designed for taxable accounts. I do not recommend it for long-term RRSP investors, nor for people with a horizon as short as 3-5 years.
There is no single solution for all index investors: a lot depends on the individual and their situation. I’m concerned, however, that there seems to be a widely held belief that dividend stocks (and ETFs) are likely to outperform broad-market ETFs over the long term simply because they have higher yield, and that a dividend strategy is optimal for investors in the accumulation phase of their lives. I don’t believe that’s the case.
Dan, thank you for the prompt answer. I have seen elsewhere that the interest for dividend paying vehicles comes from the conclusion that the market is highly unpredictable and going sideways these days, so instead of chasing growth, dividends are something you can at least count on (as they say ‘dividend stocks pay you to own them’). If I read your answer correctly, your recommendation is not to focus on yield-producing vehicles for 4-5 year horizons, but to pick one of the existing porfolios (say, Global Potato) and simply dial down the equity allocation until it becomes appropriate for the horizon (say, 30%) and keep dialing down until it goes to zero as the horizon nears.
Since people saving for a, say, house downpayment typically do it in a non-registered porfolio (TFSAs can only take $15,000 so far), don’t you agree that dividend-based index offer advantages due to dividend tax credits (as opposed paying mostly capital gain taxes)?
Perhaps this whole discussion I am trying to start is caused by the sorry state of more-less safe vehicles to for 5 year horizons (high interest accounts, bonds, GICs all pay 1-3% these days). This is causing people to consider venturing into dividend index funds, preferred shares, junk bonds etc. that dial up the risk.
@Dejan: Re: the types of investments to use for a 3-5 year rime horizon, there is no way around the fact that the price of safety is low returns. Whether you want to use dividend ETFs or more broad-based index funds probably won’t make a big difference, just always remember that a 30% to 40% decline is possible. The allocation you suggest sounds quite reasonable.
For many investors (it depends on province and income level) taxes on capital gains and dividends are not much different so check this out before letting it affect your decision.
Would a income investor buying dividends be better off using an index approach or owning the stocks outright? I’m inclined to say they should just buy the stocks. My reasons:
-A company can pay a low dividend and still make the dividend index. Income investors are usually interested in the high yield dividend payers only
-Low turnover in these portfolios, so after it’s set up, it should have maintenance fee’s lower than the MER of an ETF
So owning the stocks will result in more income and less fees. What do other people think???
I won’t dismiss dividend stock out performance as just a myth. There is evidence that breaking up portfolios into deciles based on dividend yields and looking at total returns shows that the portfolios with higher yields outperformed those with lower yields. (IIRC, you can find this data in David Dremen and Triumph of the Optimists). Having said that, I’m not convinced that (a) out performance is due to anything but the value effect (b) out performance will persist after accounting for turnover and taxes and (c) the higher returns are simply a result of higher risk (already reflected in low prices and hence higher dividend yields).
Having said that, there is no guarantee that dividend stocks will continue to outperform in the future. Especially, now that so many ETFs and income mutual funds are chasing dividend income. The unfortunate fact of capital markets is that anomalies tend to get exploited by market participants.
All of this is irrelevant if a dividend investor is picking stocks. Then, they are just a stock picker and there is plenty of evidence on how stock-pickers as a group fare (not very good).
One thing to look at when reviewing a company and its dividends are what are the sources of its dividends. There are companies that will sometimes increase debt to be able to maintain a dividend that their current revenue may not support. While the current and historical dividend payments may look good, if the company is borrowing just to be able to pay a dividend, that can be a sign of trouble.
I believe your analysis fails the survivorship bias test. Companies do go bankrupt and leave you with nothing. A long term analysis of stocks ignores this. A company giving out dividends is much more likely to be solid and less likely to disappear.
Just one other note to add to this discussion. It seems like most of you arguments here are assuming that the asset in question is held within a tax sheltered account, and therefore the tax implications of the investment are not relevant.
If you are talking about investments held outside an RRSP or TFSA, then dividends have a significant tax advantage over capital gains. (As of today, of course.) Dividend income has a tax break associated with it, since presumably the (Canadian) company in question has already paid tax on profits.
I can’t remember the exact calculation, but it works out to about just under half the amount of tax owing for a similar value in capital gains, depending on your tax bracket. Eg. if your capital gain / dividend is $5k, your tax owing on the capital gain would be ~$500, and your tax owing on the dividend is ~$250. I have a spreadsheet that calculates this, but I’m too lazy to open it.
So, for a $5k value, you get to keep $4500 vs $4750. That’s a pretty big bonus!
Especially for people who are looking for an income stream, dividends make much more sense. (And you don’t incur commissions by selling stock… Or have to sell for income when the price is low… etc.)
This means that you are entitled to 0.10 x 100 shares 10…..There are several important dates to note when a firms board of directors declares a dividend. It is important to understand that capital needs and investor expectations vary from industry to industry which is why comparison of dividends and ratios is generally most meaningful among companies within the same industry and the definition of a high or low ratio should be made within this context…..The market generally regards the cessation or reduction in dividends as bad news. It is also important to note that if a corporations board of daIRSectors declares a dividend that would come out of the corporations capital surplus or would in any way make the company insolvent some state laws may declare the dividend illegal…..The fact that dividends are a series of cash flows extending indefinitely into the future plays an important role in stock pricing.
This is far better than anything that I have written on the same subject!
I look at this very simply: Warren Buffett got started by purchasing a company and canceling it’s dividends!
Why?
So that he would have more money to invest in growing the company; investors who wanted dividends invested elsewhere … those who didn’t became multi-millionaires!
Berkshire Hathaway still operates on the same principle: why pull money out of BH when Warren can grow your money faster than you can?